IEEE Geoscience and Remote Sensing Magazine - September 2017 - 45

a consequence, the corresponding radar altimetric measurements are often discarded during quality-control checks,
even if significant efforts have been made to increase altimetric data availability when approaching land [5], [7]. But
most of the altimeters launched since the late 1980s for oceanic applications used the Ku-band frequency range, which
does not allow accurate SSH information close to the shore.
The Indo-French satellite SARAL, embarking with the
AltiKa altimeter, was launched in February 2013. It is the
first ocean altimeter to operate in the Ka band, leading to a
smaller footprint and lower noise. This frequency range is
particularly well suited for improving mesoscale and coastal observational capabilities [8], and it will be employed in
the future Surface Water and Ocean Topography (SWOT)
mission (http://swot.jpl.nasa.gov/). Results from calibration and validation analyses of the SARAL/AltiKa measurements have all demonstrated the high performance of this
altimeter mission over all surfaces, as evidenced by the
works in the 2015 special issue of Marine Geodesy [8].
While the first purpose of this article is to investigate
the value of the SARAL/AltiKa sea-level measurements, the
conservative and global approach of operational data processing has so far resulted in the systematic flagging and
rejection of altimetry data in the Solomon Sea area, preventing us from reaching this article's objective. We then
first need to establish the best method to recover as many
data as possible in the Solomon Sea.
DATA AND PROCESSING METHODOLOGY
SARAL flies on the same 35-day repeat orbit as the previous European Remote-Sensing Satellite (ERS)-1 and -2 (1991-
2003) and ENVironment SATellite (ENVISAT) (2002-2010)
missions. Compared to the Jason-2 (2008-2016), flying
on a ten-day repeat orbit, the intertrack distance is much
smaller (75 km for SARAL and 315 km for Jason-2 at the
equator), providing a relatively good spatial sampling at
the expense of a poorer time sampling.
Thirty-five SARAL and nine Jason-2 tracks cross the area
of interest. For the sake of clarity, we present here the results
for only a limited number of tracks crossing complicated
bathymetric features; they are representative of altimetry
performance in the Solomon Sea (Figure 1). Three pairs
of SARAL/Jason-2 tracks crossing the area of interest over a
long distance (~500 km) and very close in space for both
missions were chosen: Sa-361/J2-010, Sa-232/J2-251, and
Sa-060/J2-073. Tracks Sa-361/J2-010 approach near to a
Solomon Sea island at 6° S, tracks Sa-060/J2-073 cross two
Solomon Sea islands in the 6.5 to 8° S latitudinal range, and
tracks Sa-232/J2-251 pass near to a small island at 9° S and
over a lagoon with shallow waters in the 10.5-11.5° S latitudinal range.
A two-year common period (from March 2013 to February 2015) of altimeter observations is used here for both
SARAL and Jason-2, corresponding to the first 20 cycles of
the SARAL mission. Only the 20 cycles of Jason-2 that are
closest in time with SARAL/AltiKa data are analyzed (Jason-2
SEPTEMBER 2017

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

cycles 173, 176, 180, 184, 187, 191, 194, 198, 201, 205, 208,
212, 215, 219, 222, 226, 229, 233, 237, and 240).
SEA-SURFACE HEIGHT ANOMALY DATA
We use 1-Hz altimeter data from geophysical data records
(GDRs) provided by AVISO for both the SARAL and Jason-2 missions. They have a spatial resolution of 6-7 km
along the track (corresponding to measurements averaged
over 1 s), and SSHA measurements are expected with an
accuracy of 3.2 cm. All the information concerning these
products can be found in the product handbooks produced by AVISO [9], [10].
The GDR files contain the altimeter ranges (the distance
between the radar and the sea surface observed within the
altimeter footprint), a high-precision orbit (altitude), and
all relevant corrections needed to calculate the SSHA data.
The range must first be corrected from the atmospheric-related effects inducing a delay in the electromagnetic pulse
travel time (ionospheric and tropospheric effects) and from
the sea state effect, which induces an electromagnetic bias.
The corrected range is given by
corrected range = range + wet tropospheric correction
+ dry tropospheric correction
+ ionospheric correction
+ sea state bias correction.
The SSH is the height of the sea surface relative to an
arbitrary reference ellipsoid used to compute the altitude
of the satellite. It is calculated by subtracting the corrected
range from the altitude of the satellite:
SSH = altitude - corrected range.
The SSH is the sum of the height of the marine geoid and of
the dynamic topography, induced by dynamic ocean processes. Because geoid uncertainties at short wavelengths do
not allow for properly extracting the dynamic topography
from the altimetric SSH, we assume a stationary geoid and
remove an MSSH from the SSH signal to obtain the SSHA
signal. Therefore, only the temporal variability of the dynamic topography relative to a long-term mean can be estimated accurately from satellite altimeter data.
Finally, the SSHA must also be corrected from the geophysical effects of the tides, the changing atmospheric pressure (called the inverse barometer), and high-frequency wind
fluctuations:
SSHA = SSH - MSSH
- solid earth tide height
- geocentric ocean tide height
- pole tide height
- inverted barometer height correction
- high-frequency fluctuations
of the sea-surface topography.
45

http://swot.jpl.nasa.gov/

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - September 2017